Tool-heavy coding at 3B-active cost
Dense agentic coding with thinking control
Top-ranked multilingual embeddings for search
Selected Qwen models served through an optimized inference stack built for lower request cost at production volume.
Infrastructure designed for sustained throughput, stable latency, efficient GPU utilization, and cost-efficient inference.
Prompts and outputs are processed for inference, not stored after completion, and never used to train models.
Use $25 free credits to evaluate throughput, TTFT, latency, request success, and output quality before paying anything or moving traffic.
What is Qwen good for? Qwen is the broadest family of Apache-2.0 open-weight models — natively multimodal chat models, dense and sparse-MoE variants, and top-ranked multilingual embeddings — strongest in tool-heavy agentic coding and multilingual work, and priced low because its MoE models activate only about 3B parameters per request.
Qwen 3.6 35B-A3B delivers agentic coding on par with far larger models while activating just 3B parameters per token — tool-heavy loops at small-model cost.
Vision is native to the family — screenshots, PDFs, diagrams, and video frames are read alongside text, with no separate vision model to route to.
Qwen models are trained across 100+ languages, and thinking control lets one model mix instant replies with deeper handling of escalated cases.
Qwen3 Embedding 8B tops multilingual embedding leaderboards — retrieval and generation from one family, one API.
Function calling and structured output across the lineup return schema-valid JSON for ticket, PDF, and form pipelines at production volume.
3B-active MoE pricing and discounted cache reads suit offline enrichment, tagging, and report generation over large datasets.
Code examples
curl -X POST "https://api.entrim.ai/v1/chat/completions" \ -H "Content-Type: application/json" \ -H "Authorization: Bearer $ENTRIM_API_KEY" \ -d '{ "model": "Qwen/Qwen3.6-35B-A3B", "messages": [ { "role": "user", "content": "Why do they call it a building if it is already built?" } ] }'SDKs & docs
Use familiar SDK patterns with quickstart examples for common production setups.
OpenAI-compatible
Keep the request format your team already knows.
No cold starts
Production requests are served without model spin-up delays.
Prompts and outputs are processed for inference, not stored after completion, and never used to train models.
Inference and request processing stay within Entrim’s EU-operated environment, with DPA support for customers.
Requests are encrypted, access is isolated by API key, and traffic is handled with tenant-level separation.
Controls are being built around recognized security frameworks, with limited operational metadata used for billing, monitoring, latency, and abuse prevention.